oteranga bay to haywards a line (churton park … · oteranga bay to haywards reconductoring ©...

OTERANGA BAY TO HAYWARDS RECONDUCTORING © Transpower New Zealand Limited 2007. All rights reserved.

i

OTERANGA BAY TO HAYWARDS A LINE (CHURTON PARK SECTION) RECONDUCTORING

ATTACHMENT B: OPTIONS AND COSTING REPORT TRANSPOWER NEW ZEALAND LIMITED April 2018

OTERANGA BAY TO HAYWARDS RECONDUCTORING © Transpower New Zealand Limited. All rights reserved.

2

Table of Contents

1 Introduction ......................................................................................................3

1.1 Purpose .......................................................................................................3

1.2 Document Structure .....................................................................................3

2 Moving from a long list of options to a short list ................................................3

2.1 Key requirements and assessment criteria ...................................................4

2.2 Assessing optionality of short listed conductors ...........................................8

3 Short List Option Costs ....................................................................................9

3.1 Purpose .......................................................................................................9

3.2 Cost breakdown ......................................................................................... 11

3.2.1 Investigation & Design ........................................................................ 11

3.2.2 Conductors & Material Cost ................................................................ 11

3.2.3 Towers structural + foundations ......................................................... 11

3.2.4 Access & Property............................................................................... 11

3.2.5 Insulators & Hardware, Construction & other....................................... 12

3.2.6 P50 Risk Allowance ............................................................................ 12

3.2.7 Operating expenditure ......................................................................... 13

3.3 Electrical losses ......................................................................................... 13

3.4 Total present value costs ........................................................................... 14

4 Listed Project Capital Allowance .................................................................... 15

OTERANGA BAY TO HAYWARDS RECONDUCTORING © Transpower New Zealand Limited. All rights Reserved.

3

1 Introduction

This document is the Options and Costing report for the Churton Park section of the

Oteranga Bay to Haywards A line reconductoring listed project application.

The condition of the line’s conductor has reached replacement criteria and needs to

be replaced.

1.1 Purpose

The purpose of this report is to:

• explain the long list to short list process

• identify the short list options that address the identified need

• provide summarised costs for all short list options

• present our cost-benefit analysis

• explain our listed project capital allowance.

1.2 Document Structure

This report forms part of the Oteranga Bay to Haywards A line reconductoring listed

project application.

2 Moving from a long list of options to a short list

The long list of options that are alternatives to the project fall into three broad

categories:

• Non-transmission solutions or alternatives to decrease or eliminate the need for a transmission investment through the use of such things as smart metering, demand response schemes etc.

• Transmission solutions: new assets o Building a new line o Using underground cable instead of over-head lines.

• Transmission solutions: existing assets o Maintain existing asset by patch fixing o Do nothing – run to failure o Replacing the conductor on the lines and increasing the capacity o Replacing the conductor on the lines and decreasing the capacity o Replacing the conductor on the lines and keep the same capacity - the

“like for like” option o Dismantling the line and not replacing it.

Each of these long-list options has been assessed by considering their applicability to

resolving the need, the likelihood they will be cost competitive with other equivalent

options and the timeliness of the possible implementation.


4

We typically consider both transmission and non-transmission solutions (NTS) as

options to meet transmission needs. However, due to the asset condition, safety, and

asset criticality concerns associated with the existing conductor this option was not

considered further as NTS solutions were not suitable options to meet the need for

investment in this case.

2.1 Key requirements and assessment criteria

The long-list was evaluated using the following key requirements and assessment criteria:

1. Fit for purpose

• The design will meet current and forecast energy demand 2. Technically feasible

• Complexity of solution

• Reliability, availability and maintainability of the solution

• Future flexibility – fit with long term strategy for the Grid

• Ideally the design can be staged and / or have flexibility to preserve options for future changes

3. Practical to implement

• It must be possible to implement the solution by the required dates

• Implementation risks, including potential delays due to property and environmental issues

4. Good electricity industry practice (GEIP)

• Consistent with good international practice

• Ensure safety and environmental protection

• Accounts for relative size, duty, age and technological status

• Technology risks 5. Provide system security (additional benefit resulting from an economic

investment)

• Improved system security

• System operator benefits (controllability)

• Dynamic benefits (modulation features and improved system stability) 6. Indicative cost

• whether an option will clearly be more expensive than another option with similar or greater benefits

7. Feedback from consultation 8. Is economically neutral (or positive) for electricity consumers

Table 1 summarises our assessment of the long list options:


5

Table 1 - Assessment of Long List Options

Long List Short Listed

Comments

Non Transmission Alternatives

X The need (based on condition assessment and risk of conductor failure) is for a replacement conductor. As such, this option is not viable.

Transmission Solutions – New Assets

Building a new line

X

This option has been discarded. In 1992 OTB-HAY was diverted around Churton Park and consolidated into a new ‘transmission corridor’ obtained through a consent process. Therefore, it is unlikely there will be a better line route from a consenting perspective compared with the existing corridor.

Using underground cable instead of over-head lines.

X

This option has been discarded based on the cost being higher than other options. Undergrounding is very expensive compared to overhead lines. The terrain is too steep and hilly for a cable within the transmission corridor; therefore, a new route is required.

Transmission Solutions – Existing Assets

Maintain existing asset by patch fixing

X

This option has been discarded. As the conductor continues to deteriorate, our ability to effectively maintain it will reduce over time to a point where it is no longer safe or cost effective to do so. Piecemeal removal or repair of widespread defects is not practicable for this line as required access is not possible or is excessively costly in many locations due to the steep and hilly terrain and under-crossings in span.

Run to failure - wait until the conductor fails then replace either short sections or the entire line

X

This option has been discarded. This option comes with unacceptable risk to public safety. It would also result in an unplanned outage to one or both HVDC poles, which would result in major economic impacts to the electricity market.

Replacing the conductor on the lines and increasing the capacity

X

This option has been discarded. An increase in rating on the Churton Park section of the HVDC lines will not increase HVDC capacity, as its capacity is constrained by the rest of the HVDC line between Benmore and Haywards.

Replacing the conductor on the lines and decreasing the capacity

X

This option has been discarded. Decreasing the ratings of the OTB-HAY conductors would reduce the HVDC’s capacity.

Replacing the conductor on the lines and keep the same capacity - the “like for like” option

✓

This option has been included in the short list. This option meets all of our screening criteria (fit- for- purpose, technically feasible, practical, GEIP, system security, cost). A range of conductors are consistent with this option. The types of conductors that have been short listed are discussed below.

Dismantling the line and not replacing it

X This option has not been included as the there is a clear benefit provided by the HVDC to the NZ electricity system and market.


6

Reconductoring with a like-for-like or modern equivalent conductor is the only credible

option from our long-list. We have considered a range of conductors within this short-

listed option1.

The conductors selected for the short-list needed to meet the current operating

capacity of the HVDC (700 MW for Pole 3 and 500 MW for Pole 2), otherwise they

would constrain the rest of the link.

In the below table, we summarise the reconductoring options considered in order to

derive our short-list.

1 Goat, Phosphorous and Selenium duplex conductors were eliminated because to

achieve the required rating, they would exceed their recommended maximum

operating temperature.

Drake and Dublin duplex conductors are still being trialled and the results will not be

available in time for this reconductoring project


7

Table 2

Co

nd

uc

tor

Mee

ts lo

ad

gro

wth

sce

nario

s

Fu

ture

TT

U2

up

gra

de

po

ssib

le?

Characteristics of options relative to Moa ACSR Duplex

Su

itab

ility to

en

viro

n-m

en

t

Sta

nd

ard

co

nd

uc

tor

sto

ck

Sh

ort lis

t?

Moa duplex ACSR/AC (@65°C)

Yes Yes Maintains status quo. Some tower and engineering strengthening required for modern standards. Yes Yes ✓

Chukar duplex ACSR/AC (@61°C)

Yes Yes

Blowout3 slightly smaller than Moa. Loading significantly higher (20.5%CBL strung tension of conductor approx 8kN greater than that of existing Moa, and additional vertical loads from weight) meaning tower and foundation strengthening likely required. Wiring productivity slightly lower due to bigger conductor. Because of a larger conductor size, the bundle would be larger and could have structure internal clearance issues (which are already difficult to achieve for Moa) resulting in replacement cross arms with additional tower and foundation strengthening. Access costs increase substantially for cranes and concrete trucks to allow arm replacements and foundation strengthening.

Yes Yes ✓

Zebra duplex ACSR/AC (@118°C)

Yes No

Less wind load as conductor is smaller, but is lighter and will need greater tension to try to contain within easement resulting in increased likelihood of additional tower and foundation strengthening. Not all spans will be within existing easements so some expensive property easement costs are be expected. A smaller conductor means audible noise could be greater. Lighter conductor with high tensions could have internal clearance issues resulting in some replacement cross arms with additional tower and foundation strengthening. Access costs increase substantially for cranes and concrete trucks to allow arm replacements and foundation strengthening. Does not meet HVDC short term pole overload current rating.

Yes Yes ✓

Zebra triplex ACSR/AC (@65°C)

Yes Yes

Likely to have greater blowout to Moa at increased tension to stay within easement. Greater loads on tower and foundations, resulting in increased likelihood of tower and foundation strengthening. Not all spans will be within existing easements so some expensive property easement costs are be expected. A smaller conductor means audible noise could be greater - risk. Lighter conductor with high tensions could have internal clearance issues resulting in some replacement cross arms with additional tower and foundation strengthening. Access costs increase substantially for cranes and concrete trucks to allow arm replacements and foundation strengthening. Triplex bundle reduces ground clearance, requiring additional inverted V configurations. Wiring productivity is lower due to extra subconductor, sagging and space requirements.

Yes Yes ✓

Goat triplex ACSR/AC (@80°C)

Yes Yes

Likely to have greater blowout to Moa, so increased tensioning would be needed to stay within easement. Greater loads on tower and foundations, resulting in increased likelihood of tower and foundation strengthening. Not all spans will be within existing easements so some expensive property easement costs are be expected. A smaller conductor means audible noise could be greater - risk. Lighter conductor with high tensions could have internal clearance issues resulting in some replacement cross arms with additional tower and foundation strengthening. Access costs increase substantially for cranes and concrete trucks to allow arm replacements and foundation strengthening. Triplex bundle reduces ground clearance, requiring additional inverted V configurations. Wiring productivity is lower due to extra subconductor, sagging and spacering requirements.

Yes Yes ✓

Sulphur duplex AAAC/1120 (@81°C)

Yes No

Likely to have greater blowout compared to Moa. Loading could be slightly less (20.5% CBL strung tension of conductor approx 6kN less than existing Moa). Likely need for some taller tower replacements. Most spans will be outside existing easements and some substantial property easement costs should be expected. A smaller conductor means audible noise could be greater. Due to lighter tensions no major structure internal clearance issues are expected (mitigated with inverted V and existing 180kg weights). Access costs increase for cranes and concrete trucks to allow for new taller towers. Does not meet HVDC short term pole overload current rating.

Yes4 Yes ✓

2 Tactical Transmission Upgrade, which for conductors relates to thermal uprating 3 Blowout refers to the conductor moving in the wind. 4 AAAC tends to perform better in polluted environments as there is no bimetallic corrosion.


8

2.2 Assessing optionality of short listed conductors

The following table shows the operating capacities of each conductor option.

Table 3: Unquantified Assessment of Conductor Capacity

Capacity required with both

poles operating

Short term overload capacity required with pole 2 outage

Conductor Real

Power (MW)

Temp (°C)

Temp within conductor’s operating

limit?

Real Power (MW)

Temp (°C)

Temp within conductor’s operating

limit?

Moa duplex (Base) 873 65 ✓ 1008 78 ✓

Chukar duplex 883 61 ✓ 1004 71 ✓

Zebra duplex 873 118 ✓ 1001 152 X

Zebra triplex 873 65 ✓ 1006 78 ✓

Goat triplex 881 80 ✓ 1001 97 -

Sulphur duplex 878 81 ✓ 1004 99 X

The continuous capacity of the HVDC circuits is 700 MW for Pole 3 and 500 MW for

Pole 2. All conductor options meet this requirement. In a contingent event each pole

can have a short-term overload, Pole 2 up to 840 MW and Pole 3 up to 1000 MW. Both

Sulphur and Zebra duplex would exceed their recommended operating temperature if

overloaded to 1000 MW. Zebra duplex is recommended to operate at below 120°C

while Sulphur duplex at below 90°C.

If the HVDC link was further upgraded it could increase the short-term overload

capacity of Pole 2 to 1000 MW, allowing the HVDC post-contingency capacity to

increase from 840 MW to 1000 MW. This would also mean less (pre-contingency)

reserves would need to be procured by the market. However, if Sulphur or Zebra

duplex lines were installed, the post-contingent constraint in the HVDC system would

remain at 840 MW, so the lines would need to be replaced in order to realise the full

benefits from adding a fourth cable.

There is a reasonable likelihood of such an upgrade being required in the next 20 years

(ie. within the useful life of the new conductors).

Sulphur and Zebra do not meet the short term overload rating for Pole 3 now (and for

Pole 2 with a potential fourth cable) so score poorly in our unquantified benefit

“Optionality for future upgrade”.


9

The final short-list of conductors are shown in the following table:

Table 4: Short list Options

Conductor Option Type Temp (°C) MW

Moa duplex ACSR/AC 65 873

Chukar duplex ACSR/AC 61 883

Zebra duplex ACSR/AC 118 873

Zebra triplex ACSR/AC 65 873

Goat triplex ACSR/AC 80 881

Sulphur duplex AAAC/112 81 878

3 Short-List Option Costs

In this section we describe our approach to costing each of the options.

The cost of each short-listed option includes:

• the rectifications and strengthening anticipated to be required at each tower

and foundation

• an assessment of how stringing the new conductor will be carried out taking

into account aspects such as terrain, length of line, impact of circuit outages,

resources etc

• equipment and materials required to complete the works

• the extent of any ancillary work, including access tracks to tower sites,

foundation work for heavy lifting equipment, bridge strengthening (if transport

is required) and additional work required for road, rail and other utility crossings

• an assessment of the uncertainty involved in each of these aspects, for

example ground conditions and the strength of existing towers (dependant on

the steel type and condition of the foundations)

• risks of delay due to weather conditions.

3.1 Purpose

Assumptions about each of these components have been made in order to compare

options and the assessment of uncertainty is used to establish a Listed Project Capex

Allowance (LPCA).

The various risks associated with each of these elements are described through each

section.

To determine which conductors to evaluate in the short-list we started by considering

the cost of a wide variety of conductors. Our cost estimates for the base case (Moa

duplex) considered the following:


10

• Transpower Enterprise Estimating System (TEES) costings for construction

using specific conductor types, and

• Transpower Business Case estimates derived from a high-level desk top study

for duplex Moa installation (being the Base case), with an uncertainty range of

-50%/+50%. The capital costs included:

o Investigation and design

o Materials (conductors, insulators & hardware)

o Construction (conductors, structures, foundations, access and

property)

For other short-list conductor options, material cost estimates were obtained for each,

as were high level estimates of tower and foundation strengthening costs. All other

construction costs were assumed to be the same as for Moa duplex installation. This

may tend to favour options other than Moa duplex due to potential additional property

costs, which we have considered within our unquantified benefit assessment.

A more accurate “Solution Study Report” (SSR) was subsequently undertaken for the

preferred option (Moa). We have adjusted the previous estimated costs for the other

conductor options to reflect this new cost information5. The new SSR Moa cost was

approximately $5.5 million higher than the “old” cost estimate and can be seen in Table

5. As can be seen this change has been predominantly due to an increase in Access

and Property costs which are common to all options. We have rescoped and priced

the other short-listed conductor options considering their relative cost and scoping

differences to Moa (as outlined in Table 2) and these ‘new’ costs have been used

throughout this final listed project application proposal.

Table 5 – Changes to costs through project

Capex, real 2018, $000 SSR Cost

March 2018

Desktop Cost

December 2017

Change

Investigation 0 544 -544 -100%

Design 876 1,270 -393 -31%

Conductor materials cost 915 959 -44 -5%

Insulators & Hardware 369 397 -28 -7%

Tower structural + foundations 405 1,040 -635 -61%

Access & property 7,372 923 6,450 699%

Stringing & other construction 9,959 10,221 -262 -3%

P50 risk allowance 1,858 0 1,858 0%

Reserve costs 0 900 -900 -100%

Total capex 21,754 16,254 5,500 34%

5 New cost option A = Adjusted cost of Moa SSR based on estimated scope variation (ref Table 2) extrapolated from the loading and clearance information of the Moa SSR and input from our costing models.


11

3.2 Cost breakdown

3.2.1 Investigation & Design

Investigation costs are costs related to identifying our preferred solution and

developing this proposal. Although these are a true project cost, they are common

across all options and have already been allocated to Transpower as part of our current

RCP2 Base Capex Allowance. Design costs are being included here, which cover the

costs of detailed design and technical investigations and studies to implement the

preferred solution.

3.2.2 Conductors & Material Cost

The cost of the conductor itself differs between the different options but the work

involved to string the new conductor is common between the options. For example,

the costs and materials are higher for options that are triplex.

3.2.3 Towers structural + foundations

Some tower and foundation strengthening is required for this project to meet the

structural loads using current tower modelling techniques and foundation data. There

are small changes required to increase clearances for safe working or for statutory

requirements (e.g. NZECP 34). Most other options require a quantum of strengthening

as tensions increase, with the larger Chukar conductor requiring all towers and

foundations to be strengthened. A small number of tower replacements could also be

required as well as cross-arm replacement to meet these heavier loads.

3.2.4 Access & Property

Most of the line section to be replaced passes over farmland with none directly over

urban buildings, although four spans are less than 100m from houses in Churton Park.

The crossing over the busy SH 1 Johnsonville-Porirua motorway and the electrified

North Island Main Trunk Railway at span 58A - 59A pose a significant safety risk. There

are also a number of significant crossings over 11 kV and 33 kV supply lines and minor

roads. We have worked with the distribution asset owners to relocate or underground

line crossings where possible. Replacing the conductor over the remaining crossings

will require careful management to ensure public safety. Protection of these crossings

will involve construction of safety-nets and supporting structures with dismantling

following the stringing of the new conductor. Such crossings are both labour and time

intensive.

We have planned for the use of scaffolding hurdles for the major State Highway 1 and

Electrified NIMT crossings. This is because Catenary Support Systems are not yet

sufficiently mature to be used with the much larger Moa conductor and duplex

configuration in such a long span.

Costs to access the transmission line involve utilisation of existing access tracks, with

some upgrades required to meet the loading of large stringing equipment and cranes


12

used during construction. Each wiring site will have a flat platform constructed for

wiring machines and conductor storage. These and temporary access tracks will be

removed at completion – these costs have increased from our preliminary estimates

since the SSR work has been undertaken. Construction will be challenging due to the

hilly nature of the terrain, long access routes and multiple crossings. These costs

include earthworks, benching, track upgrades, obtaining associated consents and

landowner permissions and any reinstatement works required.

The work falls within allowable activities under the National Environmental Standards

for Electricity Transmission (NES)6 and Electricity Act. We also have some property

issues that need to be rectified and we have included some funds to negotiate a

resolution.

3.2.5 Insulators & Hardware, Construction & other

These cost categories capture all other major costs, construction costs such as

stringing costs (the labour and associated tools and machinery hire), and insulators

and hardware required to be replaced in order to upgrade the new conductor. These

costs do vary a little across the different conductor types due to the slightly different

work required on some options. This category also contains a small contingency to

allow for weather delays – ie. wind speeds of greater than 80 km/hour will curtail work

and >40 km/hour make sagging impossible. Analysis of local wind data suggests these

wind conditions will prevail around 12% of the time, although this is variable year to

year. The outages have been timed to coincide with the lowest wind months to

minimise the adverse weather effects.

3.2.6 P50 Risk Allowance

As detailed design has not occurred for the options yet, there is a risk associated with

the P50 estimate. This cost category accounts for additional tower strengthening and

foundation work as well as stringing costs that are going to be encountered through

the detailed design stage and implementation of our preferred option. The full extent

of tower strengthening is subject to detailed design. The outage window is very tight

with completion planned on the day before Easter Friday. Any delay during execution

of the work will push the programme out past Easter.

6 The NES sets out a national framework of permissions and consent requirements for activities on existing electricity transmission lines. Activities include the operation, maintenance and upgrading of existing lines.


13

Table 6: Capital costs ($2018, 000s)

Capex, real $2018, 000s

Moa duplex Chukar duplex

Zebra duplex

Zebra triplex Goat triplex Sulphur duplex

Investigation & Design

876 876 876 876 876 876

Conductor material cost

915 1,063 627 883 769 659

Insulators & Hardware

369 369 369 369 369 369

Towers + foundations

405 2,898 1,209 2,898 2,898 2,345

Access & Property

7,372 7,997 7,537 7,997 7,997 7,572

Stringing + other construction

9,959 10,725 10,068 12,158 12,140 10,260

P50 risk allowance

1,858 1,858 1,858 1,858 1,858 1,858

Total P50 cost 21,754 25,786 22,544 27,039 26,907 23,939

3.2.7 Operating expenditure

We have assumed operating costs of $400k per annum, which is based on the average

spend on this section of the line over the last 3 years. We don’t expect there to be any

material differences in the operating costs across the short-list options.

3.3 Electrical losses

In addition to the capital costs we have also considered the potential benefits resulting

from lower electrical losses.

There are differences in the losses from each of the conductors. Larger conductors that run at lower temperatures will result in lower electrical losses. We have estimated the losses for each conductor under the five MBIE 2016 EDGS7 scenarios:

1. Mixed renewables

2. High Grid

3. Global Low Carbon

4. Disruptive

5. Tiwai off

We have used SDDP8 – a hydro-thermal dispatch optimisation model – to estimate

flows on the HVDC under a range of hydrological conditions. SDDP takes 78 years of

historical hydro inflow data and produces an optimal hydro dispatch profile given future

demand, fuel/carbon price, and generation plant scenarios.

7 Electricity Demand and Generation Scenarios 8 Stochastic Dual Dynamic Programming


14

We have considered potential losses over 40 years using our “P50” expected demand

forecast. These have been valued at $100/MWh and discounted using a 7% discount

rate to determine the present value of losses associated with each option.

We found that in the Mixed renewables scenario northward transfers averaged around

2200 GWh in 2020, reducing to 1600GWh by 2040. In the “Tiwai off” scenario they

averaged just over 6200 GWh in 2020, slowly reducing to around 4900 GWh by 20409.

For all scenarios, we took the average losses (from all the 78 inflow years).

We valued these losses using three different price assumptions:

• The short run marginal cost (SRMC) derived from our SDDP market model

• $50 per MWh sensitivity

• $150 per MWh sensitivity.

Table 7 shows the present value of the losses when averaged across the five EDGS scenarios, using a 7% pa discount rate. The expected life of the asset was assumed to be 40 years for valuing the losses.

Zebra duplex has the highest losses, while Chukar has the lowest.

Table 7: Present value of losses, average of 5 EDGS scenarios ($000)

PV $000 Moa

duplex Chukar duplex

Zebra duplex

Zebra triplex

Goat triplex

Sulphur duplex

$50 sensitivity 1,296 1,172 2,427 1,618 2,061 1,669

SRMC 2,546 2,303 4,768 3,179 4,050 3,216

$150 sensitivity 3,888 3,516 7,281 4,854 6,184 5,008

3.4 Total present value costs

The following table summarises the capital and operating costs, and also shows the

present value (PV) of these costs.

Table 8: Conductor cost comparisons (P50 estimates)

$2018, 000s Moa


Zebra duplex

Zebra triplex

Goat triplex

Sulphur duplex

Capital cost 21,754 25,786 22,544 27,039 26,907 23,939

Annual opex (over life of asset)

400 400 400 400 400 400

Total present value (PV) cost 25,219 28,813 25,923 29,930 29,812 27,167

9 Assuming that all lower South Island transmission constraints are alleviated.


15

The following table shows our overall analysis of options, including both costs and

electrical losses.

Table 9: PV Costs and Losses showing preferred option

Option Moa


Zebra duplex

Zebra triplex

Goat triplex

Sulphur duplex

Total present value cost

25,219

28,813

25,923

29,930

29,812

27,167

Total present value losses

2,546

2,303

4,768

3,179

4,050

3,216

Total present value costs + losses

27,765

31,116

30,691

33,109

33,862

30,383

Net Benefit vs Base Case - -3,351 -2,926 -5,344 -6,097 -2,618

Rank 1 4 3 5 6 2

4 Listed Project Capital Allowance

Transpower is seeking approval from the Commission to increase the Base Capex

Allowance by the estimated Listed Project Capital Allowance of the application.

Transpower estimates the expected cost of the application to be $21.8 million (+ HVDC

Reserve Costs) in current (2018) dollars. With the addition of inflation and financing

costs the total cost becomes $23.5 million (plus HVDC Reserve Costs) in 2020 when

the conductor replacement is completed.

We have derived our proposed LPCA in a manner consistent with it being a standalone

project, on the basis that our existing Base Capex Allowance was approved for other

works, not including this project.

A summary of our LPCA calculation, including financing costs, inflation and exchange

rate uncertainty (but excluding HVDC Reserve Costs) is shown in Table 10 and in

Table 11 the annual break down is shown. As shown, the total LPCA we are applying

for is $23.5 million. It is important to recognise that this amount excludes HVDC

Reserve Costs.

We consider this amount to be our P50 estimate of the cost of the project – that is there

is an equal chance that the project could be delivered for more or could be delivered

for less. As with any project, and consistent with the incentive regime, we will attempt

to deliver this project as efficiently as possible.


16

Table 10: Derivation of Listed Project Capex Allowance

LPCA application

Point selected

within distribution

(probability)

Cost applied for

($000)

Capex (real 2018$) P50 21,754

Inflation 758

Exchange rates -

IDC 952

Total LPCA (2020$) 23,464

Table 10: Listed Project Capex Allowance Annual Allocation

Cost by year 2018 2019 2020

Capex (real 2018$) 584 4,181 16,989

Inflation 1 93 664

Exchange rates - - -

IDC 20 115 817

Total LPCA (2020$) 605 4,389 18,471

Per span costs The costs of this project are higher than previous reconductoring works when

compared on a per-span basis. These higher costs can be attributed to:

• the size and weight of the conductor and fittings when compared to typical

Zebra conductor

• multiple wiring crews working at the same time to mitigate the length of the

outage

• abnormally difficult terrain to work in

• a property easement remediation

• difficult and expensive hurdle crossings over SH1 and the electrified main trunk

railway

• short 3km wiring runs which are not as efficient as a 6km run, and

• expensive undergrounding of local distribution company lines.

When these additions are accounted for, the project is comparable in cost with other

reconductoring works we have completed.


17

5 HVDC Reserve costs

5.1 How reserve costs are allocated to Transpower

In the electricity market, reserves are required to protect against a sudden failure of a

large generating plant or the HVDC link. This service is required to stop the resulting

fall in frequency and allow the system frequency to recover promptly to 50 Hz.

Reserves are provided by generation, or interruptible load. Reserve costs are paid by

asset owners of generating units greater than 60 MW and the HVDC owner (being

Transpower as the asset owner).

Costs are allocated on an island basis, proportional to the quantity of electricity injected

by a generator or the HVDC transfer quantity.10 With both poles in service

Transpower’s allocation of the reserve costs is reduced due to the ability for each pole

to cover an outage of the other pole.

A simplified representation of the allocation of reserve costs is set out below for

illustration11:

𝑆ℎ𝑎𝑟𝑒 𝑜𝑓 𝑟𝑒𝑠𝑒𝑟𝑣𝑒𝑠𝑡 = 𝑟𝑒𝑠𝑒𝑟𝑣𝑒 𝑐𝑜𝑠𝑡𝑡 ×𝐻𝑉𝐷𝐶 𝑟𝑖𝑠𝑘 ,𝑡 − 30 𝑀𝑊ℎ

𝑡𝑜𝑡𝑎𝑙 𝑟𝑒𝑠𝑒𝑟𝑣𝑒 𝑟𝑒𝑞𝑢𝑖𝑟𝑒𝑚𝑒𝑛𝑡𝑡

where:

𝐻𝑉𝐷𝐶 𝑟𝑖𝑠𝑘 ,𝑡 is the at risk HVDC transfer in trading period t

𝑡𝑜𝑡𝑎𝑙 𝑟𝑒𝑠𝑒𝑟𝑣𝑒 𝑟𝑒𝑞𝑢𝑖𝑟𝑒𝑚𝑒𝑛𝑡𝑡 is the total reserve requirement including that

required from transmission and from generation units.

When both poles are in service 𝐻𝑉𝐷𝐶 𝑟𝑖𝑠𝑘 ,𝑡 is significantly lower than the flow on the

HVDC as one pole has the ability to cover the outage of the other pole reducing HVDC

transfer at risk. When we are running a monopole, there is no self-coverage and all of

the HVDC transfer is at risk. As a result, our allocation of the HVDC reserve costs will

increase significantly as a result of undertaking this work.

5.2 How we modelled the impact on Transpower reserve

costs

The same SDDP runs that were used to model the various outage options (see

Attachment D) have been used to help analyse the potential impact on Transpower

reserve costs. This provided generation, HVDC flows and short-run marginal cost of

generation (SRMC) information.

However, SDDP is a least-cost optimisation, and the SRMC it produces is not

necessarily the same as the market spot prices that may occur. For example,

generators may be more risk adverse in a dry summer if there is a risk of low lake

10 See clause 8.59 in the Code for details 11 Note that this simple representation is for illustration purposes only and excludes some additional terms. See 8.59 for the full details.


18

levels leading into winter, resulting in higher summertime spot prices (than the SRMC

of SDDP may imply). This is exactly the behaviour we observed this last summer in

January 2018.

In order to better capture these market dynamics, we have supplemented the SDDP

SRMC model outputs with a separate monte carlo simulation which produces a richer

range of pricing outcomes. The parameters for the “shape” of spot prices over the year

have been set so that it is consistent with the price patterns observed in the historical

data.

We have assumed that reserve prices are a function of spot prices, and this

relationship has been determined using least squares regression techniques. Volatility

in reserve prices (that is unrelated to spot prices movements) is reflected in the Monte

Carlo simulation.

We have also included in our simulation the potential impact of an unplanned CCGT

outage (during the HVDC outage). We assume that weekly average spot prices would

rise to at least $75/$125/$200 per MWh in a wet/normal/dry year, and that if the thermal

outage occurred during the HVDC outage then prices would increase a further $50 per

MWh. These assumptions reflect the type of marginal thermal plant that may be

operating under each scenario.

5.3 The impact of outage on Transpower reserve costs

We intend to recover and capitalise the HVDC reserve costs as part of this project.

However, the extent of these costs is heavily dependent on hydrological conditions. In

wet years the flows on the HVDC are likely to be higher such that our allocation of the

share of reserves will be higher. It is also likely less thermal generation plant will be

operating which again is likely to increase our allocation of the total reserve costs. Our

modelling suggests that the increase in Transpower reserve costs could be as low as

$11 thousand or as large as $6 million with a 50th percentile of $1.9 million.

Table 11 summarises the range of reserve cost increases that we may be exposed to

in different hydrological years12. It shows the increase in costs when there is just one

pole operating, compared to the cost when both poles are operating. A negative

number means that the share of costs has reduced under monopole operation.

We have excluded 10 days of VBE testing from our calculation, since those

incremental costs cannot be attributed to this reconducting project.

Note that it is still possible for Transpower to be exposed to higher reserve costs (than

we have modelled) if an extreme market event results in greater reserve market

impacts than we have assumed. For example, we assume that in a dry year an

appropriate price floor is $250 per MWh (weekly price), if there is an unplanned thermal

12 In a wet year, the reserve cost per MWh will tend to be lower, however Transpower’s share of the costs will be much higher due to the higher volume of HVDC transfers North. In a dry year the reserve cost per MWh will tend to be higher, so costs increase for all parties (both Transpower and generators).


19

outage during the HVDC outage. Its possible that a security of supply risk produces

spot prices in excess of this for the week, resulting in very high reserve prices.

Table 11- Impact of outage on Reserve cost shares

Percentile Transpower share $m Generator share $m Total

market $m

Transpower % share

Mean 1,983 -355 1,628 122%

0% 11 422 433 2%

1% 499 128 627 80%

10% 908 -16 892 102%

50% 1,862 -433 1,428 130%

90% 3,142 -454 2,688 117%

99% 4,476 -491 3,986 112%

100% 5,954 -1,289 4,665 128%

Approving a P50 cost would leave us with a significant risk that we have little control

over and few options to mitigate. There is no forward market for HVDC reserves to use

to hedge our exposure. A partial mitigation option would have been to construct a by-

pass line to reduce the outage length. However, as explained in this application this

is not feasible within the timeframes for the need of this project, and would not be

economic to implement.

5.4 Treatment of HVDC Reserve Costs within our Application

Given the high level of dependence on hydrology associated with reserve costs that is

beyond our control, we consider that they should not be considered within the incentive

regime (i.e. Base Capex expenditure adjustment).

This could be facilitated through use of the g term in Schedule B, Division 1 of the

Capex IM. The Base Capex expenditure adjustment can be represented as:

𝑏𝑎𝑠𝑒 𝑐𝑎𝑝𝑒𝑥 𝑒𝑥𝑝𝑒𝑛𝑑𝑖𝑡𝑢𝑟𝑒 𝑎𝑑𝑗𝑢𝑠𝑡𝑚𝑒𝑛𝑡 = 𝑎 × (𝑏 − 𝑐 − 𝑔)

where:

a is the Base Capex incentive rate (33%)

b is the adjusted Base Capex Allowance

c is the actual Base Capex cost capitalised

g is the net Base Capex for which the incentive does not apply.


20

For example, if our listed project is approved, the approved amount will increase the

“b” term. The actual cost of the project will be capitalised and included in the “c” term,

and would include any increase in reserve costs. If our reserve costs were not included

in our proposed increase to our Base Capex Allowance (i.e. the “b” term”) but instead

captured in the “g” term to offset their appearance in the c term, then they would “net-

out” and not impact on the Base Capex expenditure adjustment.

In this case we would not be penalised if the weather was such that we faced high

reserve costs or stand to gain if we faced very low reserve costs. Given this

arrangement we would expect the reserve costs to be excluded from the approved

increase to the Base Capex Allowance as they would not be subject to the incentive

calculations.

The actual costs associated with reserves and stand-down costs would be captured in

the c term for this project but then could be subtracted out using the g term such that

they net out and are removed from the Base Capex expenditure adjustment.

Table 12 shows the range of incremental reserve costs that could occur, across a set

of 78 historical inflow years. Either wet or dry hydro conditions could cause Transpower

reserve costs to increase above the 90th percentile. We propose that once the hydro

conditions in 2020 have transpired, the reserve cost impact is recalculated using the

“actual” hydro conditions.

Table 12 – Transpower reserve cost risk

Percentile 0% 10% 50% 90% 100%

Increased cost (real 2018$) 11 908 1,862 3,142 5,954

Inflation 0 37 76 128 243

IDC 0 22 46 77 146

Total contingency for reserve costs (2020$) 11 967 1,983 3,348 6,344

oteranga bay to haywards a line (churton park … · oteranga bay to haywards reconductoring ©...

Documents